Presented by Apptio
AI has become one of the largest categories of technology spending yet just 37% of AI initiatives are delivering their expected business value. For technology and finance leaders, the challenge is no longer simply identifying promising AI use cases. It’s determining which investments are creating enough value to justify their cost — and which investments to start, stop, optimize or scale.
That requires moving beyond project-by-project decisions to a management model that treats AI investments as a portfolio focused on business outcomes.
How AI economics builds business value into AI strategies
Understanding what an AI initiative truly costs is often more complicated than looking at a cloud invoice, software license or model usage charge. Infrastructure, model consumption such as tokens, data management, governance, security, operational support and labor all contribute to the true cost of ownership.
AI economics — the relationship between the total cost of ownership for AI and the business value those investments create — gives technology and finance leaders a way to evaluate those investments consistently. Rather than comparing costs, consumption or adoption across disparate use cases, organizations can evaluate cost per outcome and determine where additional investment will generate the greatest return.
Shifting from project management to IT portfolio management
That visibility becomes particularly important as AI investments compete for funding against other AI projects, cybersecurity, data analytics and business intelligence, process automation, public cloud, digital transformation and technical debt. Visibility across the technology portfolio supports higher-level, more strategic conversations and decisions on AI investments.
Traditionally, we have seen that technology governance asks whether an individual project is on budget. AI economics should be asking whether the portfolio is creating value.
This opens the door for leaders to answer a consistent set of questions:
-
What is the business spending?
-
Who is driving that spend?
-
Which initiatives are delivering measurable value?
-
Where should additional investment be directed?
-
Which initiatives should be reconsidered?
No single executive typically has the complete picture. CIOs and CTOs provide visibility into technology consumption and capacity; CFOs bring investment discipline and financial accountability; and business leaders define outcomes and value realization.
Making those comparisons requires connecting financial, operational, consumption and organizational data that often resides in separate systems and can be interpreted differently by different functions. Creating that common view allows leaders to move the conversation from “How much are we spending?” to “What are we delivering to the business?”
IBM’s own CIO organization offers an example of this approach in practice. By using the Apptio integrated suite to connect financial, IT and business views, the organization gained greater transparency into application total cost of ownership and how resources aligned with strategic initiatives. That shared view helped teams make more informed decisions about application criticality, consumption and spend, while identifying opportunities to reinvest in transformation priorities.
Establishing start, stop, optimize and scale parameters
A practical framework sorts AI initiatives using two questions: Is the business value real and measured against a baseline? And is the cost per outcome lower than the way the work was done before?
The answers determine whether an initiative should start, stop, optimize or scale.
Start: Before funding a new initiative, leaders should have a defined outcome metric, a baseline against which to measure it and clear criteria for stopping it. They should also determine whether an existing capability can solve the same problem.
Stop: If the expected business outcome hasn’t been achieved within the agreed timeframe, adoption has stalled or cost per outcome can’t be brought below the baseline, it may be time to halt the initiative. Stopping underperforming projects frees resources for investments with greater potential.
Optimize: An initiative may be creating demonstrable business value while still costing too much per workflow or outcome. Rather than eliminating it, leaders can examine model selection, utilization, infrastructure or design to improve its economics before scaling. This distinction matters: without visibility into both cost and value, organizations risk cutting promising AI capabilities simply because their economics haven’t yet been optimized.
Scale: Initiatives that consistently demonstrate measurable value at sustainable economics are candidates for additional investment and organizational support. That value might take the form of productivity gains, faster delivery, increased throughput, better customer outcomes or revenue growth.
Together, these criteria help create a continuous optimization cycle that allows the C-suite to reallocate resources based on evidence rather than enthusiasm.
Managing AI as an investment portfolio
As AI spending grows, leaders need to be able to explain not only what they are spending, but what the organization is receiving in return. A common view of cost, consumption and business outcomes gives technology, finance and business leaders a shared basis for making those decisions.
Organizations that develop that capability can identify which investments deserve more funding, which require optimization and which should be stopped — while giving boards a clearer view of how AI spending translates into business value.
This kind of portfolio-level decision-making depends on bringing together information that typically sits across technology, financial, operational, usage and organizational systems. IBM approaches that through IBM Apptio, applying technology financial intelligence, grounded in FinOps practices and IT financial management (ITFM), to connect cost, ownership, consumption and unit economics with the value an investment produces.
The objective is not simply greater visibility into AI spending. It’s to give technology, finance and business leaders the shared context to decide which AI investments warrant additional funding, further optimization or cutting.
The future of AI leadership will not be defined by who spends the most, but by who understands the economics of those investments well enough to know where to spend next.
Take control of AI investments with AI Investment Management.
William Lobig is IBM Vice President, Apptio.
Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact sales@venturebeat.com.